INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 thrott

-0.71

 tradem

-0.70

ĸļ

-0.66

 pregnancies

-0.65

 redundancy

-0.65

 twins

-0.63

 hinges

-0.63

 carriers

-0.60

 acceleration

-0.60

 pige

-0.59

POSITIVE LOGITS

sic

0.96

yer

0.73

ashed

0.73

met

0.71

NRS

0.69

jac

0.68

fixed

0.68

elson

0.68

imen

0.67

issue

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.